This master´s thesis deals with calibration and determination of uncertainties of measurement for focus variation microscope Alicona Infinite Focus G4. Together with calibration, a measurement of chosen parameters with calculation of measurement uncertainties of a component was realised. The thesis also describes the whole calibration process and terminology used in calibration, as well as the calibrated instrument, it´s parameters and usage in practice. The measurements were realized on premises of Intemac Solutions s.r.o. Obtained data were than processed using MS Excel, Gwyddion and TalyMap software. At the end of the thesis, practical recommendations are formulated.

The goal of the work is to find suitable soulition for a small water plant in kontinuity of a projekt of damaged dam repair.

Mount Cameroon is considered the only active volcano along a 1600 km long chain of volcanic complexes called the Cameroon Volcanic Line (CVL). It has erupted seven times during the last 100 years, the most recent was in May 2000. The approximately 500,000 inhabitants that live and work around the fertile flanks are exposed to impending threats from volcanic eruptions and earthquakes. In this thesis, a hazard assessment study that involves both statistical modelling of seismic hazard parameters and the evaluation of a future volcanic risk was undertaken on Mount Cameroon. The Gutenberg-Richter magnitude-frequency relations, the annual activity rate, the maximum magnitude, the rate of volcanic eruptions and risks assessment were examined. The seismic hazard parameters were estimated using the Maximum Likelihood Method on the basis of a procedure which combines seismic data containing incomplete files of large historical events with complete files of short periods of observations. A homogenous Poisson distribution model was applied to previous recorded volcanic eruptions of Mount Cameroon to determine the frequency of eruption and assess the probability of a future eruption. Frequency-magnitude plots indicated that Gutenberg-Richter b-values are partially dependent on the maximum regional magnitude and the method used in their calculation. b-values showed temporal and spatial variation with an average value of 1.53 ± 0.02. The intrusion of a magma body generating the occurrence of relatively small earthquakes as observed in our instrumental catalogue, could be responsible for this high anomalous b-value. An epicentre map of locally recorded earthquakes revealed that the southeastern zone is the most seismically active part of the volcano. The annual mean activity rate of the seismicity strongly depends on the time span of the seismic catalogue and results showed that on average, one earthquake event occurs every 10 days. The maximum regional magnitude values which had been determined from various approaches overlap when their standard deviations are taken into account. However, the magnitude distribution model of the Mt. Cameroon earthquakes might not follow the form of the Gutenberg-Richter frequency magnitude relationship. The datations of the last eruptive events that have occurred on Mt. Cameroon volcanic complex are presented. No specific pattern was observed on the frequency of eruptions, which means that a homogenous Poisson distribution provides a suitable model to estimate the rate of occurrence of volcanic eruptions and evaluate the risk of a future eruption. Two different approaches were used to estimate the mean eruption rate (λ) and both yielded a value of 0.074. The results showed that eruptions take place on average once every 13 years and, with the last eruption occurring over 15 years ago, it is considered that there is at present a high risk of an eruption to occur. / Dissertation (MSc)--University of Pretoria, 2016. / Geology / MSc / Unrestricted

An enormous and urgent energy demand is predicted due to the growing global population, increase in power intensive industries, higher living standards, electrification of remote areas, and globalisation (transportation). Moreover, the global consciousness about the harmful effects of traditional methods of power generation on the environment. That, in turn, has created a need to strategically plan and develop renewable and sustainable energy generation systems. This study presents a wind resource assessment of seven locations proximate to the largest industrial hub in the Middle East, Jubail Industrial City, Kingdom of Saudi Arabia, and a Geographic Information System, GIS based model considering a multi-criteria wind farm site suitability approach for the entire Kingdom of Saudi Arabia and elsewhere. The hourly mean wind speed data at 10, 50 and 90 m above the ground level (AGL) over a period of five years was used for a meteorological station at the Industrial Area (Central) of Jubail. At the remaining six sites, the meteorological data were recorded at 10 m AGL only. Five years of wind data were used for five sites and three years of data were available for the remaining one site. At the Industrial Area (East), the mean wind speeds were found to be 3.34, 4.79 and 5.35 m/s at 10, 50 and 90 m AGL, respectively. At 50 and 90 m AGL, the availability of wind speed above 3.5 m/s was more than 75%. The local wind shear exponent, calculated using measured wind speed values at three heights, was found to be 0.217. The mean wind power density values at measurement heights were 50.92, 116.03 and 168.46 W/m2, respectively. After the assessment and comparison of wind characteristics of all seven sites, the highest annual mean wind speed of 4.52 m/s was observed at Industrial Area (East) and the lowest of 2.52 m/s at the Pearl Beach with standard deviations of 2.52 and 1.1 m/s, respectively. In general, at all sites, the highest monthly mean wind speed was observed in February/June and the lowest in September/October. The period of higher wind availability coincides with a high power demand period in the region attributable to the air conditioning load. The wind rose plots show that the prevailing wind direction for all sites was from the north-west. Weibull parameters for all sites were estimated using maximum likelihood, least-squares regression method (LSRM), and WAsP algorithm. In general, at all sites, the Weibull parameter, c, was the highest in the months of February/June and the lowest in the month of October. The most probable and maximum energy carrying wind speed was determined by all three methods. The highest value of most probable wind speed was found to be in the range of 3.2 m/s to 3.6 m/s at Industrial Area (East) and the highest value of maximum energy carrying wind speed was found to be in the range 8.6 m/s to 9.0 m/s at Industrial Area 2 (South) by three estimation methods. The correlation coefficient (R2), root mean square error (RMSE), mean bias error (MBE), and mean bias absolute error (MAE) showed that all three methods represent wind data at all sites accurately. However, the maximum likelihood method is slightly better than LSRM, followed by WAsP algorithm. The wind power output at all seven sites, from five commercially available wind turbines of rated power ranging from 1.8 to 3.3 MW, showed that Industrial Area (East) is most promising for wind farm development. At all sites, based on percentage plant capacity factor, PCF, the 1.8 MW wind turbine was found to be the most efficient. At Industrial Area (East), this wind turbine was found to have a maximum PCF of 41.8%, producing 6,589 MWh/year energy output. The second best wind turbine was 3 MW at all locations except the Al-Bahar Desalination Plant and Pearl Beach. At both of these locations, 3.3 MW was the next best option. The energy output from the 3 MW wind turbine at Industrial Area (East) was found to be 11,136 MWh/year with a PCF of 41.3%. The maximum duration of rated power output from all selected wind turbines was observed to be between 8 to 16.6% at Industrial Area 2 (South). The minimum duration of rated power output, less than 0.3% for all wind turbines, was observed at Pearl Beach. The maximum duration of zero power output of between 35 to 60% was also observed at Pearl Beach. / Thesis (PhD)--University of Pretoria, 2017. / Mechanical and Aeronautical Engineering / PhD / Unrestricted

Philosophiae Doctor - PhD / One of the assumptions of conventional regression analysis is I that the parameters are constant over all observations. It has often been suggested that this may not be a valid assumption to make, particularly if the econometric model is to be used for economic forecasting0 Apart from this it is also found that econometric models, in particular, are used to investigate the underlying interrelationships of the system under consideration in order to understand and to explain relevant phenomena in structural analysis. The pre-requisite of such use of econometrics is that the regression parameters of the model is assumed to be constant over time or across different crosssectional units.

When developing an aircraft, one of several important aspects is to predict and properly design the dynamic behaviour of the aircraft. This holds for manned aircraft as well as for UAVs. The optimal dynamic behaviour for an aircraft depends on the mission or purpose: for a certain use an aircraft should be agile, other may require a more stable one. In aeronautics, the properties that describe the aircraft ecacy with respect to some task are known as ying qualities, and our goal is to study their dependence on some design parameters. As a test model we use an existing UAV. After deriving its 6-DOF dynamic model and assessing its baseline characteristics, we perform parametric studies. The strategy followed is divided in two steps: the rst consists on analyzing ying qualities sensitivity to changes in model parameters. The second step studies how specific design changes affect model parameters. Because the rst step only depends on the dynamic model form, we verify, by testing two other dierent aircrafts, that conclusions drawn from this step are valid to other congurations. Finally we show how results from parametric studies can be used to improve the UAV ying qualities regarding a certain mission, through the introduction of slight modications on baseline design.

As the world is moving towards a more digital world with the majority of people having tablets, smartphones and smart objects, solving real-world computational problems with handheld devices seems more common. Detection or tracking of objects using a camera is starting to be used in all kinds of fields, from self-driving cars, sorting items to x-rays, referenced in Introduction. Object detection is very calculation heavy which is why a good computer is necessary for it to work relatively fast. Object detection using lightweight models is not as accurate as a heavyweight model because the model trades accuracy for inference to work relatively fast on such devices. As handheld devices get more powerful and people have better access to object detection models that can work on limited-computing devices, the ability to build their own small object detection machines at home or at work increases substantially. Knowing what kind of factors that have a big impact on object detection can help the user to design or choose the correct model. This study aims to explore what kind of impact distance, angle and light have on Inceptionv2 SSD, MobileNetv3 Large SSD and MobileNetv3 Small SSD on the COCO dataset. The results indicate that distance is the most dominant factor on the Inceptionv2 SSD model using the COCO dataset. The data for the MobileNetv3 SSD models indicate that the angle might have the biggest impact on these models but the data is too inconclusive to say that with certainty. With the knowledge of knowing what kind of factors that affect a certain model’s performance the most, the user can make a more informed choice to their field of use.

The world’s energy use in buildings (residential and commercial) accounts for around 40% of the worldwide energy consumption, and space heating is the responsible for half of the energy need in the building sector. In Europe, only a small share (less than 10%) of existing buildings was built after 1990. Most of the building stock does not satisfy the recent energy technical standards; in addition there is a very low trend to construct new buildings in the last years. Renovation of the existing buildings is a feasible option to reduce the energy need in Europe, but finding the optimum solutions for a renovation is not a simple task. Each design parameter differently influences the final energy need of buildings and, furthermore, the different variables are differently correlated each other. Building refurbishment will benefit from a tool for the selection of the best measures in term of energy need. This work, through a global sensitivity analysis, aims at determining the contribution of the design parameters to the building energy demand and the correlation between the different variables. The considered parameters are related to the improvement of the thermal transmittance of both the opaque envelope and the windows, the solar transmittance of the glazing surfaces, the window size, the thermal inertia of the internal walls and the external sunshades for windows. Several dynamic simulations have been performed varying the design parameters from different starting conditions. Finally, due to the large number of cases elaborated, an inferential statistical analysis has been performed in order to identify the predominant factors and the correlation between the design parameters in a global context.

Ignition delay time, τ_ign, is a key quantity of interest that is used to assess the predictability of a chemical kinetic model. This dissertation explores the sensitivity of τ_ign to uncertainties in: 1. rate-rule kinetic rates parameters and 2. enthalpies and entropies of fuel and fuel radicals using global and local sensitivity approaches. We begin by considering variability in τ_ign to uncertainty in rate parameters. We consider a 30-dimensional stochastic germ in which each random variable is associated with one reaction class, and build a surrogate model for τ_ign using polynomial chaos expansions. The adaptive pseudo-spectral projection technique is used for this purpose. First-order and total-order sensitivity indices characterizing the dependence of τ_ign on uncertain inputs are estimated. Results indicate that τ_ign is mostly sensitive to variations in four dominant reaction classes. Next, we develop a thermodynamic class approach to study variability in τ_ign of n-butanol due to uncertainty in thermodynamic properties of species of interest, and to define associated uncertainty ranges. A global sensitivity analysis is performed, again using surrogates constructed using an adaptive pseudo-spectral method. Results indicate that the variability of τ_ign is dominated by uncertainties in the classes associated with peroxy and hydroperoxide radicals. We also perform a combined sensitivity analysis of uncertainty in kinetic rates and thermodynamic properties which revealed that uncertainties in thermodynamic properties can induce variabilities in ignition delay time that are as large as those associated with kinetic rate uncertainties. In the last part, we develop a tangent linear approximation (TLA) to estimate the sensitivity of τ_ign with respect to individual rate parameters and thermodynamic properties in detailed chemical mechanisms. Attention is focused on a gas mixture reacting under adiabatic, constant-volume conditions. The proposed approach is based on integrating the linearized system of equations governing the evolution of the partial derivatives of the state vector with respect to individual random variables, and a linearized approximation is developed to relate ignition delay sensitivity to scaled partial derivatives of temperature. The computations indicate that TLA leads to robust local sensitivity predictions at a computational cost that is order-of-magnitude smaller than that incurred by finite-difference approaches.

>Magister Scientiae - MSc / This research objectively investigates the e ectiveness of machine learning (ML) tools towards predicting several geo-physical parameters. This is based on a large number of studies that have reported high levels of prediction success using ML in the eld. Therefore, several widely used ML tools coupled with a number of di erent feature sets are used to predict six geophysical parameters namely rainfall, groundwater, evapora- tion, humidity, temperature, and wind. The results of the research indicate that: a) a large number of related studies in the eld are prone to speci c pitfalls that lead to over-estimated results in favour of ML tools; b) the use of gaussian mixture models as global features can provide a higher accuracy compared to other local feature sets; c) ML never outperform simple statistically-based estimators on highly-seasonal parame- ters, and providing error bars is key to objectively evaluating the relative performance of the ML tools used; and d) ML tools can be e ective for parameters that are slow- changing such as groundwater.